Many nonnormalities, one simulation: Do different data generation algorithms affect study results?

Monte Carlo simulation studies are among the primary scientific outputs contributed by methodologists, guiding application of various statistical tools in practice. Although methodological researchers routinely extend simulation study findings through follow-up work, few studies are ever replicated. Simulation studies are susceptible to factors that can contribute to replicability failures, however. This paper sought to conduct a meta-scientific study by replicating one highly cited simulation study (Curran et al., Psychological Methods, 1, 16-29, 1996) that investigated the robustness of normal theory maximum likelihood (ML)-based chi-square fit statistics under multivariate nonnormality. We further examined the generalizability of the original study findings across different nonnormal data generation algorithms. Our replication results were generally consistent with original findings, but we discerned several differences. Our generalizability results were more mixed. Only two results observed under the original data generation algorithm held completely across other algorithms examined. One of the most striking findings we observed was that results associated with the independent generator (IG) data generation algorithm vastly differed from other procedures examined and suggested that ML was robust to nonnormality for the particular factor model used in the simulation. Findings point to the reality that extant methodological recommendations may not be universally valid in contexts where multiple data generation algorithms exist for a given data characteristic. We recommend that researchers consider multiple approaches to generating a specific data or model characteristic (when more than one is available) to optimize the generalizability of simulation results.

One step at a time: A statistical approach for distinguishing mediators, confounders, and colliders using direction dependence analysis

In observational data, understanding the causal link when estimating the causal effect of an independent variable (x) on a dependent variable (y) often requires researchers to identify the role of a third variable in the x → y relationship. Mediation, confounding, and colliding are three key third-variable effects that yield different theoretical and methodological implications for drawing causal conclusions. Commonly used covariance-based statistical methods, such as linear regression and structural equation modeling, cannot distinguish these effects in practice, however. In this study, we introduce a statistical approach for distinguishing mediators, confounders, colliders, and potential M-bias structures that uses higher-order moment information from the data. We propose a two-step procedure that uses the Hilbert-Schmidt independence criterion within the direction dependence analysis framework. Results from Monte Carlo simulations show that our proposed approach accurately recovers the true data-generating process of the third variable. We provide an empirical example to demonstrate the application of our proposed approach in psychological research. Finally, we discuss implications and future directions of our work. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Design, analysis, and interpretation of treatment response heterogeneity in personalized nutrition and obesity treatment research

It is increasingly assumed that there is no one-size-fits-all approach to dietary recommendations for the management and treatment of chronic diseases such as obesity. This phenomenon that not all individuals respond uniformly to a given treatment has become an area of research interest given the rise of personalized and precision medicine. To conduct, interpret, and disseminate this research rigorously and with scientific accuracy, however, requires an understanding of treatment response heterogeneity. Here, we define treatment response heterogeneity as it relates to clinical trials, provide statistical guidance for measuring treatment response heterogeneity, and highlight study designs that can quantify treatment response heterogeneity in nutrition and obesity research. Our goal is to educate nutrition and obesity researchers in how to correctly identify and consider treatment response heterogeneity when analyzing data and interpreting results, leading to rigorous and accurate advancements in the field of personalized medicine.

Low normal FMR1 genotype in older adult women: Psychological well-being and motor function

The FMR1 gene plays a key role in adult neurogenesis and neuroplasticity, and thus may contribute to age-related health in the population. The current study focused on the "low normal" FMR1 genotype, defined by lower-than-typical numbers of FMR1 CGG repeats (<26), as a potential genetic determinant of age-related health. We characterized the effect of the low normal FMR1 genotype on psychological well-being and motor function in a racially diverse non-clinical sample of older adult women. Women with low CGG repeats were distinguished from those with CGGs falling within the mid-high end of the normal range by reduced performance on multimodal assessments of motor function and psychological well-being, with large effect sizes. Robust continuous associations were also detected between lower CGG repeat length and reduced psychological well-being, balance, and dexterity. Findings suggest that FMR1 may represent an important mediator of individual differences in age-related health; larger epidemiological studies are needed. Given that approximately 23-35% of females carry the low normal genotype, efforts to understand its clinical effects have relevance a broad swath of the aging population.

The role of frailty in the association between depression and fall risk among older adults

OBJECTIVES

Although there is a recognized association between depression and greater fall risk among older adults, the mechanisms explaining this association are unclear. This study evaluated the role of frailty, a common geriatric syndrome, in determining greater risk of falls among older adults with depression.

METHOD

We used longitudinal data from three biennial waves of the Health and Retirement Study (HRS; 2010-2014). The sample included community-dwelling survey respondents age ≥ 65 who participated in objective physiological measures. Major Depression (MD) was measured using Composite International Diagnostic Interview for depression short form. Frailty was measured using criteria outlined in the frailty phenotype model. Causal mediation analysis was used to differentiate the direct effect of depression and indirect effect mediated by frailty on falls, fall injuries, and multiple falls.

RESULTS

Major depression was associated with significantly greater odds of experiencing a fall (OR: 1.91; 95% CI: 1.31, 2.77), fall injury (OR: 1.86; 95% CI: 1.17, 2.95), and multiple falls (OR: 2.26; 95% CI: 1.52, 3.37) over a two-year period. Frailty was a significant mediator of the effects of depression on falls and multiple falls, accounting for approximately 18.9% and 21.3% of the total effects, respectively. We found no evidence of depression-frailty interaction. Sensitivity analyses showed that results were robust to unmeasured confounding and alternative operationalizations of depression.

CONCLUSION

Frailty explains a significant proportion of increased likelihood of falls among older adults with depression. Treatment and management of frailty symptoms may be an important components of fall prevention among older adults with depression.

Future Orientation Among Children Affected by Parental HIV in China: An Exploratory Analysis of Complex Interactions

We utilized an exploratory analytic approach to examine predictors of children's future beliefs, an internal asset associated with resilience among children affected by HIV, with emphasis on complex interactions among multisystem factors. Children (N = 1221) affected by parental HIV in China reported on psychosocial functioning, as well as internal, familial, and community resilience assets. Exploratory data analysis was conducted using a binary segmentation program. Six binary splits on predictors accounted for 22.78% of the variance in future expectation, suggesting interactions between children's perceived control of their future, loneliness, caregiver trust, and social support. Four binary splits accounted for 23.15% of the variance in future orientation, suggesting multiway interactions between control of the future, loneliness, social support, and perceived stigma. Findings suggest combinations of resilience factors are associated with children's positive future beliefs. Implications for screening, prevention, and intervention among Chinese children affected by parental HIV are discussed.

A Guideline for Reporting Mediation Analyses of Randomized Trials and Observational Studies: The AGReMA Statement

Importance

Mediation analyses of randomized trials and observational studies can generate evidence about the mechanisms by which interventions and exposures may influence health outcomes. Publications of mediation analyses are increasing, but the quality of their reporting is suboptimal.

Objective

To develop international, consensus-based guidance for the reporting of mediation analyses of randomized trials and observational studies (A Guideline for Reporting Mediation Analyses; AGReMA).

Design, Setting, and Participants

The AGReMA statement was developed using the Enhancing Quality and Transparency of Health Research (EQUATOR) methodological framework for developing reporting guidelines. The guideline development process included (1) an overview of systematic reviews to assess the need for a reporting guideline; (2) review of systematic reviews of relevant evidence on reporting mediation analyses; (3) conducting a Delphi survey with panel members that included methodologists, statisticians, clinical trialists, epidemiologists, psychologists, applied clinical researchers, clinicians, implementation scientists, evidence synthesis experts, representatives from the EQUATOR Network, and journal editors (n = 19; June-November 2019); (4) having a consensus meeting (n = 15; April 28-29, 2020); and (5) conducting a 4-week external review and pilot test that included methodologists and potential users of AGReMA (n = 21; November 2020).

Results

A previously reported overview of 54 systematic reviews of mediation studies demonstrated the need for a reporting guideline. Thirty-three potential reporting items were identified from 3 systematic reviews of mediation studies. Over 3 rounds, the Delphi panelists ranked the importance of these items, provided 60 qualitative comments for item refinement and prioritization, and suggested new items for consideration. All items were reviewed during a 2-day consensus meeting and participants agreed on a 25-item AGReMA statement for studies in which mediation analyses are the primary focus and a 9-item short-form AGReMA statement for studies in which mediation analyses are a secondary focus. These checklists were externally reviewed and pilot tested by 21 expert methodologists and potential users, which led to minor adjustments and consolidation of the checklists.

Conclusions and Relevance

The AGReMA statement provides recommendations for reporting primary and secondary mediation analyses of randomized trials and observational studies. Improved reporting of studies that use mediation analyses could facilitate peer review and help produce publications that are complete, accurate, transparent, and reproducible.

Antidepressant Use Partially Mediates the Association Between Depression and Risk of Falls and Fall Injuries Among Older Adults

BACKGROUND

The association between depression and fall risk in older adults is recognized, yet the mechanisms underlying this association are unclear. This study estimated the mediating role of antidepressant use in the association between depression and falls and fall injuries.

METHODS

Longitudinal data from the Health and Retirement Study (2004-2006) were linked with medication data from the Prescription Drug Study (2005). The sample included community-dwelling adults aged ≥65 with data on depression and medication use (n = 3565). Depression was measured using 2 independent survey tools: Composite International Diagnostic Interview for depression short form and an 8-item version of the Center for Epidemiological Studies-Depression scale. We used causal mediation analysis to estimate and compare the direct and indirect (mediated by antidepressant use) effects of depression on falls and fall injuries.

RESULTS

Individuals with major depressive disorder were significantly more likely to experience a fall (OR: 1.92; 95% CI: 1.41, 2.62) and a fall injury (OR: 1.67; 95% CI: 1.09, 2.55) over 2 years. Indirect effect estimates showed that antidepressant medication use accounted for approximately 19% and 18% of the association between major depressive disorder and falls and fall injuries, respectively. Results were similar when using an alternative depression measure and when considering only selective serotonin reuptake inhibitor antidepressants.

CONCLUSIONS

Antidepressant use explains a significant proportion, but not a majority, of the association between depression and greater fall risk. Treatment benefits of antidepressants should be considered with, and may outweigh, concerns about increased risk of falls associated with antidepressant use.

Pharmacokinetics and Pharmacodynamics of Immediate- and Modified-Release Mycophenolic Acid Preparations in Healthy Beagle Dogs

Background: Mycophenolic acid (MPA) is a broad-acting immunomodulating agent that may be therapeutically beneficial for the treatment of immune-mediated diseases in canine patients.

Objectives: To determine the suppressive effects of MPA on T-cell proliferation, and to assess the feasibility of a canine-specific q24 h modified-release MPA formulation (OKV-1001b).

Animals: Fifteen healthy purpose-bred male beagle dogs.

Methods: Two nearly identical open-label fifteen-day studies were conducted in which dogs were randomized to receive mycophenolate mofetil (MMF; 10 mg/kg q12h), or two doses of OKV-1001b (270 mg and 180 mg; q24h). Serial pharmacokinetic (PK) and pharmacodynamic (PD) samples were collected on Days 1, 8, and 15. MPA plasma concentrations were measured by liquid chromatography-tandem mass spectrometry (LC-MS/MS), while an ex vivo T-cell proliferation assay assessed PD effects. Dogs were continuously monitored for evidence of side effects and gastrointestinal tolerability.

Results: MPA induced inhibition of T-cell proliferation was observed following administration of all MPA preparations in a clear concentration-dependent manner. The PK/PD relationship was maintained across all days and time-points. Data generated herein suggest that MPA plasma concentrations above 600 ng/mL achieve at least 50% inhibition of T-cell proliferation.

Conclusions and Clinical Importance: MPA holds therapeutic potential for treating dogs with immune-mediated disease, but clinical trials will be necessary to determine its safety and efficacy in naturally occurring disease. Likewise, q24h oral modified release MPA preparations that maintain MPA plasma concentrations between 600 and 1,000 ng/mL are warranted for further studies in client-owned dogs.

A multi-stop time-of-flight spectrometer for the measurement of positron annihilation-induced electrons in coincidence with the Doppler-shifted annihilation gamma photon

In this study, we describe an advanced multi-functional, variable-energy positron beam system capable of measuring the energies of multiple "positron-induced" electrons in coincidence with the Doppler-shifted gamma photon resulting from the annihilation of the correlated positron. The measurements were carried out using the unique characteristics of the digital time-of-flight spectrometer and the gamma spectrometer available with the advanced positron beam system. These measurements have resulted in (i) the first digital time-of-flight spectrum of positron annihilation-induced Auger electrons generated using coincident signals from a high-purity Ge detector and a micro-channel plate, (ii) a two-dimensional array of the energy of Doppler-broadened annihilation gamma and the time-of-flight of positron-annihilation induced Auger electrons/secondary electrons measured in coincidence with the annihilation gamma photon, and (iii) the time-of-flight spectra of multiple secondary electrons ejected from a bilayer graphene surface as a result of the impact and/or annihilation of positrons. The novelty of the gamma-electron coincidence spectroscopy has been demonstrated by extracting the Doppler-broadened spectrum of gamma photons emitted due to the annihilation of positrons exclusively with 1s electrons of carbon. The width of the extracted Doppler-broadened gamma spectrum has been found to be consistent with the expected broadening of the annihilation gamma spectrum due to the momentum of the 1s electrons in carbon.

Fitting Ordinal Factor Analysis Models With Missing Data: A Comparison Between Pairwise Deletion and Multiple Imputation

This study compares two missing data procedures in the context of ordinal factor analysis models: pairwise deletion (PD; the default setting in Mplus) and multiple imputation (MI). We examine which procedure demonstrates parameter estimates and model fit indices closer to those of complete data. The performance of PD and MI are compared under a wide range of conditions, including number of response categories, sample size, percent of missingness, and degree of model misfit. Results indicate that both PD and MI yield parameter estimates similar to those from analysis of complete data under conditions where the data are missing completely at random (MCAR). When the data are missing at random (MAR), PD parameter estimates are shown to be severely biased across parameter combinations in the study. When the percentage of missingness is less than 50%, MI yields parameter estimates that are similar to results from complete data. However, the fit indices (i.e., χ2, RMSEA, and WRMR) yield estimates that suggested a worse fit than results observed in complete data. We recommend that applied researchers use MI when fitting ordinal factor models with missing data. We further recommend interpreting model fit based on the TLI and CFI incremental fit indices.

An Empirical Mediation Analysis of Mechanisms Underlying HIV-1-Associated Neurocognitive Disorders

HIV-1-associated neurocognitive disorders (HAND), characterized by alterations in the core components of cognitive function and age-related disease progression, persist in the post-cART era. However, the neurobehavioral mechanisms that mediate alterations in the core components of cognitive function and the progression of neurocognitive impairments have yet to be systematically evaluated. To address this knowledge gap, statistical mediation analysis was assessed, providing a critical opportunity to empirically evaluate putative neurobehavioral mechanisms underlying HAND. Neurocognitive assessments, conducted in HIV-1 transgenic (Tg) and control animals across the functional lifespan (i.e., Postnatal Day (PD) 30 to PD 600), tapped multiple cognitive domains including preattentive processes, learning, sustained attention, and long-term episodic memory. Three longitudinal mediation models were utilized to assess whether deficits in preattentive processes mediate alterations in learning, sustained attention and/or long-term episodic memory over time. Preattentive processes partially mediated the relationship between genotype and learning, genotype and sustained attention, and genotype and long-term episodic memory across the functional lifespan, explaining between 44% and 58% of the HIV-1 transgene effect. Understanding the neurobehavioral mechanisms mediating alterations in HAND may provide key targets for the development of a diagnostic biomarker, novel therapeutics, and cure/restoration strategies.

Investigating Construct Validity of the Cyber-Peer Experiences Questionnaire

Abstract. Cyberaggression and cybervictimization have gained momentum as a research focus given associated mental health sequelae. To date, however, there remains little consensus on the conceptualization and measurement of these constructs. The purpose of this study was to explore construct validity evidence for the Cyber-Peer Experiences Questionnaire (C-PEQ), a novel measure that assesses experiences of cyberaggression and cybervictimization. Undergraduate participants ( n = 735) completed an online survey including the C-PEQ and other theoretically related instruments. Confirmatory factor analysis did not provide support for the hypothesized two-factor model [MLM χ2(125) = 293.58, RMSEA = .06, CFI = .87, SRMR = .06]. The C-PEQ displayed evidence for internal consistency reliability. Evidence for convergent validity with theoretically similar constructs was mixed. Specific areas of model misspecification included items 1 and 2 on both subscales (altering social media friend lists). Future research may explore additional construct validity evidence of the C-PEQ in novel samples with these items removed.

Assessing Acceptability of the Term: "Psychopathology" Among Youth Aged 18-25

A prevailing model for mental health care for youth and families is to provide services within a "psychopathology" focused framework. This approach can compound problems for youth by imparting negative labels on them, and may be associated with iatrogenic impacts of interventions (e.g., stigmatization, lowered self-efficacy, dependency). This study assessed perceptions of the term "psychopathology" among 486 youth aged 18-25, with 39% of these youth receiving prior mental health services. Results indicated statistically significant differences in perception of the term, with youth who had received mental health services perceiving it more negatively than youth who had not. Findings suggest receipt of mental health services among young people may sensitize them to negative aspects of the term psychopathology, indicating the need for caution in using this term and other terms that may have negative impacts on mental health service use among youth.

Vagal Tone as a Putative Mechanism for Pragmatic Competence: An Investigation of Carriers of the FMR1 Premutation

Pragmatic language skills exist across a continuum in typical and clinical populations, and are impaired in many neurodevelopmental disorders, most notably autism. The mechanisms underlying pragmatic impairment are poorly understood, although theory suggests dampened vagal tone plays a role. This study investigated the FMR1 premutation as a genetic model that may lend insight into the relationship between vagal function and pragmatic ability. Participants included 38 women with the FMR1 premutation and 23 controls. Vagal tone accounted for significant variance in pragmatics across both groups and statistically mediated the effect of FMR1 premutation status on pragmatic ability. Results support vagal tone as a biophysiological correlate of pragmatic ability, which informs potential mechanistic underpinnings and could have implications for targeted treatment.

Biobehavioral composite of social aspects of anxiety in young adults with fragile X syndrome contrasted to autism spectrum disorder

Social anxiety is a common disorder that has negative impacts across multiple domains of function. Several clinical groups are at elevated risk for social anxiety, including those with fragile X syndrome and those with autism spectrum disorder. Measuring social anxiety in these clinical subgroups is fraught with challenge, however, given the complexity of social anxiety and measurement limitations that are particularly acute in persons with neurodevelopmental disorders. The over-arching aim of this study was to contribute to our understanding of the nature of social anxiety in fragile X syndrome and its association with autism spectrum disorder. To address this aim, we created a multi-faceted composite representing behavioral and biological aspects of social anxiety and examined differences in two adolescent and young adult-aged groups: 59 males with fragile X syndrome and 18 males with autism spectrum disorder. Results indicated a lower score on the multivariate composite for the males with fragile X syndrome relative to autism spectrum disorder but with evidence that traits of autism and social anxiety overlap. We conclude that measuring anxiety and autism traits in fragile X syndrome and autism spectrum disorder is complex with features that overlap and interact in a dynamic manner.

Sex Matters: Robust Sex Differences in Signal Detection in the HIV-1 Transgenic Rat

Sex differences in human immunodeficiency virus type-1 (HIV-1) have been repeatedly suggested. Females, who account for 51% of HIV-1 seropositive individuals, are inadequately represented in clinical and preclinical studies, as well as in the description of HIV-1 associated neurocognitive disorders (HAND). Direct comparisons of neurocognitive decline in women and men must be made to address this underrepresentation. The effect of biological sex (i.e., the biological factors, including chromosomes and hormones, determining male or female characteristics; WHO, 2017) on sustained attention, which is commonly impaired in HIV-1 seropositive individuals, was investigated in intact HIV-1 transgenic (Tg) and control animals using a signal detection operant task. Analyses revealed a robust sex difference in the rate of task acquisition, collapsed across genotype, with female animals meeting criteria in shaping (at least 60 reinforcers for three consecutive or five non-consecutive sessions) and signal detection (70% accuracy for five consecutive or seven non-consecutive sessions) significantly more slowly than male animals. Presence of the HIV-1 transgene also had a significant effect on shaping and signal detection acquisition, with HIV-1 Tg animals displaying significant deficits in the rate of acquisition relative to control animals–deficits that were more prominent in female HIV-1 Tg animals. Once the animals’ reached asymptotic performance in the signal detection task, female animals achieved a lower percent accuracy across test sessions and exhibited a decreased response rate relative to male animals, although there was no compelling evidence for any effect of transgene. Results indicate that the factor of biological sex may be a moderator of the influence of the HIV-1 transgene on signal detection. Understanding the impact of biological sex on neurocognitive deficits in HIV-1 is crucial for the development of sex-based therapeutics and cure strategies.

Evaluating Behavioral Health Interventions for Military-Connected Youth: A Systematic Review

INTRODUCTION

Military-connected youth are at increased risk for experiencing distinct psychosocial and behavioral health vulnerabilities. Although behavioral health interventions have been developed to treat vulnerabilities in military-connected youth, little is known about the methodological quality of studies evaluating these interventions. In this study, a systematic review of behavioral health interventions for military-connected youth was conducted to examine methodological quality and treatment outcomes.

MATERIALS AND METHODS

Electronic databases were systematically searched for studies evaluating behavioral health interventions for military-connected youth which yielded 3,324 citations. Methodological quality was evaluated by 2 researchers with 3 measures that assessed scientific rigor, transparency, external and internal validity, and power for quantitative, qualitative, and mixed-method trials. Interrater reliability was strong (κ = 0.81). Sample characteristics and treatment outcomes were also assessed.

RESULTS

Fourteen studies meeting full inclusion criteria evaluated 10 behavioral health interventions. Methodological quality scores for all studies were poor to fair, with limitations in reporting, external and internal validity, and power. Research designs were predominantly nonexperimental. Treatment effects for both psychosocial and behavioral health outcomes were consistently positive for all studies. In studies reporting effect sizes, treatment effects were small to moderate (d = 0.01-0.42, odds ratio = 0.04-0.47, b = -0.02-0.56). Demographic and military characteristics of samples were inconsistently reported.

CONCLUSION

Behavioral health interventions for military-connected youth have noteworthy methodological limitations, indicating a need to employ more rigorous research strategies. Positive treatment outcomes, however, suggest promising interventions for improving psychosocial and behavioral health problems in military-connected youth. Future research directions and implications for clinical-community practice are also discussed.

Visual Antipriming Effect: Evidence from Chinese Character Identification

Marsolek et al. (2006) have differentiated antipriming effects from priming effects, by adopting a novel priming paradigm comprised of four phases that include a baseline measurement. The general concept of antipriming supports the overlapping representation theory of knowledge. This study extended examination of the Marsolek et al. (2006) paradigm by investigating antipriming and priming effects in a series of Chinese character identification tasks. Results showed that identification accuracy of old characters was significantly higher than baseline measurements (i.e., the priming effect), while identification accuracy of novel characters was significantly lower than baseline measurements (i.e., the antipriming effect). This study demonstrates for the first time the effect of visual antipriming in Chinese character identification. It further provides new evidence for the overlapping representation theory of knowledge, and supports generalizability of the phenomenon to Chinese characters.

Altered sensitivity to social gaze in the FMR1 premutation and pragmatic language competence

Background

The FMR1 premutation affects 1:291 women and is associated with a range of cognitive, affective, and physical health complications, including deficits in pragmatic language (i.e., social language). This study investigated attention to eye gaze as a fundamental social-cognitive skill that may be impaired in the FMR1 premutation and could underlie pragmatic deficits. Given the high prevalence of the FMR1 premutation, efforts to define its phenotype and mechanistic underpinnings have significant public health implications.

Methods

Thirty-five women with the FMR1 premutation and 20 control women completed an eye-tracking paradigm that recorded time spent dwelling within the eye region in response to a face displaying either direct or averted gaze. Pragmatic language ability was coded from a conversational sample using the Pragmatic Rating Scale.

Results

Women with the FMR1 premutation failed to show attentional preference to direct gaze and spent more time dwelling on the averted eyes relative to controls. While dwelling on the eyes was associated with better pragmatic language performance in controls, these variables were unrelated in the women with the FMR1 premutation.

Conclusions

Altered sensitivity to social gaze, characterized by increased salience of averted gaze, was observed among women with the FMR1 premutation. Furthermore, women with the FMR1 premutation were unable to capitalize on information conveyed through the eyes to enhance social-communicative engagement, which differed from patterns seen in controls. These findings contribute to the growing characterization of social and communication phenotypes associated with the FMR1 premutation.

Best (but oft-forgotten) practices: mediation analysis

This contribution in the "Best (but Oft-Forgotten) Practices" series considers mediation analysis. A mediator (sometimes referred to as an intermediate variable, surrogate endpoint, or intermediate endpoint) is a third variable that explains how or why ≥2 other variables relate in a putative causal pathway. The current article discusses mediation analysis with the ultimate intention of helping nutrition researchers to clarify the rationale for examining mediation, avoid common pitfalls when using the model, and conduct well-informed analyses that can contribute to improving causal inference in evaluations of underlying mechanisms of effects on nutrition-related behavioral and health outcomes. We give specific attention to underevaluated limitations inherent in common approaches to mediation. In addition, we discuss how to conduct a power analysis for mediation models and offer an applied example to demonstrate mediation analysis. Finally, we provide an example write-up of mediation analysis results as a model for applied researchers.

Mediators of Social Connection in a Group Teleconference Intervention

BACKGROUND AND PURPOSE

Social connection has a positive influence on cancer survivorship and has been targeted in intervention work. We examined whether the formation of social connection is influenced by fear, fatalism, and isolation.

METHODS

We conducted a mediation analysis on data from an intervention study with African American breast cancer survivors to investigate whether fear, fatalism, and isolation conveyed effects on social connection.

RESULTS

Although results indicated that there were no significant mediated effects of the intervention through these variables, we garnered information about action and conceptual theories underlying the program that inform future work.

IMPLICATIONS FOR PRACTICE

Our insights can be considered in future interventions conducted for African American women with breast cancer during and following treatment.

A randomized controlled trial of a resilience-based intervention on psychosocial well-being of children affected by HIV/AIDS: Effects at 6- and 12-month follow-up

BACKGROUND

Global literature suggests that resilience-based interventions may yield improvements in psychosocial well-being for vulnerable children, but limited data are available regarding the efficacy of such interventions among children affected by parental HIV/AIDS.

OBJECTIVE

To evaluate initial efficacy of a multi-level resilience-based intervention among children affected by parental HIV/AIDS in China in improving children's psychosocial well-being and resilience-related outcomes.

METHOD

Seven hundred-ninety children, 6-17 years of age, were recruited from rural China. Children were either AIDS orphans or were living with one or two parents infected with HIV/AIDS. Children and primary caregivers were randomly assigned to participate in a 4-arm trial to evaluate the Child-Caregiver-Advocacy Resilience (ChildCARE) intervention. This resilience-based psychosocial intervention provides programming at three levels (child, caregiver, community). Survey data were collected at baseline, 6-months, and 12-months in order to examine efficacy of the child-only and child + caregiver arms in improving children's psychological resilience.

RESULTS

Intervention groups displayed improvements in several resilience-related outcomes at 6- and 12-month follow-ups, including self-reported coping, hopefulness, emotional regulation, and self-control. The child-only intervention arm showed some fading of intervention effects by 12-months.

CONCLUSION

Preliminary findings suggest that the ChildCARE intervention is efficacious in promoting psychosocial well-being of children affected by parental HIV/AIDS in rural China. Targeting both children and caregivers for psychosocial intervention may be effective in improving children's resilience. Additional evaluation and modifications, including the inclusion of booster sessions, should be considered to further strengthen the program.

Marital conflict in early childhood and adolescent disordered eating: emotional insecurity about the marital relationship as an explanatory mechanism

Disordered eating behaviors, including frequent dieting, unhealthy weight control behaviors (e.g., vomiting and skipping meals for weight loss) and binge eating are prevalent among adolescents. While negative, conflict-ridden family environments have long been implicated as problematic and a contributing factor to the development of disordered eating, few studies have examined the influence of marital conflict exposure in childhood to understand the development of these behaviors in adolescence. The current study investigates the impact of marital conflict, children's emotional insecurity about the marital relationship, and disordered eating behaviors in early adolescence in a prospective, longitudinal study of a community sample of 236 families in Midwest and Northeast regions of the U.S. Full structural mediation analyses utilizing robust latent constructs of marital conflict and emotional insecurity about the marital relationship, support children's emotional insecurity as an explanatory mechanism for the influence of marital conflict on adolescent disordered eating behaviors. Findings are discussed with important implications for the long-term impact of marital conflict and the development of disordered eating in adolescence.

The influence of race and comorbidity on the timely initiation of antiretroviral therapy among older persons living with HIV/AIDS

OBJECTIVES

We examined whether the timely initiation of antiretroviral therapy (ART) differed by race and comorbidity among older (≥ 50 years) people living with HIV/AIDS (PLWHA).

METHODS

We conducted frequency and descriptive statistics analysis to characterize our sample, which we drew from 2005-2007 Medicaid claims data from 14 states. We employed univariate and multivariable Cox regression analyses to evaluate the relationship between race, comorbidity, and timely ART initiation (≤ 90 days post-HIV/AIDS diagnosis).

RESULTS

Approximately half of the participants did not commence ART promptly. After we adjusted for covariates, we found that older PLWHA who reported a comorbidity were 40% (95% confidence interval = 0.26, 0.61) as likely to commence ART promptly. We found no racial differences in the timely initiation of ART among older PLWHA.

CONCLUSIONS

Comorbidities affect timely ART initiation in older PLWHA. Older PLWHA may benefit from integrating and coordinating HIV care with care for other comorbidities and the development of ART treatment guidelines specific to older PLWHA. Consistent Medicaid coverage helps ensure consistent access to HIV treatment and care and may eliminate racial disparities in timely ART initiation among older PLWHA.

Improving Our Ability to Evaluate Underlying Mechanisms of Behavioral Onset and Other Event Occurrence Outcomes: A Discrete-Time Survival Mediation Model

The purpose of this article is to introduce and describe a statistical model that researchers can use to evaluate underlying mechanisms of behavioral onset and other event occurrence outcomes. Specifically, the article develops a framework for estimating mediation effects with outcomes measured in discrete-time epochs by integrating the statistical mediation model with discrete-time survival analysis. The methodology has the potential to help strengthen health research by targeting prevention and intervention work more effectively as well as by improving our understanding of discretized periods of risk. The model is applied to an existing longitudinal data set to demonstrate its use, and programming code is provided to facilitate its implementation.

A Bayesian Approach for Estimating Mediation Effects with Missing Data

Methodologists have developed mediation analysis techniques for a broad range of substantive applications, yet methods for estimating mediating mechanisms with missing data have been understudied. This study outlined a general Bayesian missing data handling approach that can accommodate mediation analyses with any number of manifest variables. Computer simulation studies showed that the Bayesian approach produced frequentist coverage rates and power estimates that were comparable to those of maximum likelihood with the bias-corrected bootstrap. We share a SAS macro that implements Bayesian estimation and use two data analysis examples to demonstrate its use.

The PHLAME (Promoting Healthy Lifestyles: Alternative Models' Effects) firefighter study: testing mediating mechanisms

This paper examines the mechanisms by which PHLAME (Promoting Healthy Lifestyles: Alternative Models' Effects), a health promotion intervention, improved healthy eating and exercise behavior among firefighters, a population at high risk for health problems due to occupational hazards. In a randomized trial, 397 firefighters participated in either the PHLAME team intervention with their work shift or a control condition. Intervention sessions taught benefits of a healthy diet and regular exercise, and sought to improve social norms and social support from coworkers for healthy behavior. At posttest, team intervention participants had increased their fruit and vegetable consumption as compared to control participants. An increase in knowledge of fruit and vegetable benefits and improved dietary coworker norms partially mediated these effects. Exercise habits and VO2 max were related to targeted mediators but were not significantly changed by the team intervention. Partial support was found for both the action and conceptual theories underlying the intervention. Our findings illustrate how an effective program's process can be deconstructed to understand the underpinnings of behavior change and refine interventions. Further, fire stations may improve the health of firefighters by emphasizing the benefits of healthy diet and exercise behaviors while also encouraging behavior change by coworkers as a whole.

Evaluating mediation and moderation effects in school psychology: a presentation of methods and review of current practice

Third variable effects elucidate the relation between two other variables, and can describe why they are related or under what conditions they are related. This article demonstrates methods to analyze two third-variable effects: moderation and mediation. The utility of examining moderation and mediation effects in school psychology is described and current use of the analyses in applied school psychology research is reviewed and evaluated. Proper statistical methods to test the effects are presented, and different effect size measures for the models are provided. Extensions of the basic moderator and mediator models are also described.

A general model for testing mediation and moderation effects

This paper describes methods for testing mediation and moderation effects in a dataset, both together and separately. Investigations of this kind are especially valuable in prevention research to obtain information on the process by which a program achieves its effects and whether the program is effective for subgroups of individuals. A general model that simultaneously estimates mediation and moderation effects is presented, and the utility of combining the effects into a single model is described. Possible effects of interest in the model are explained, as are statistical methods to assess these effects. The methods are further illustrated in a hypothetical prevention program example.

A general model for testing mediation and moderation effects

This paper describes methods for testing mediation and moderation effects in a dataset, both together and separately. Investigations of this kind are especially valuable in prevention research to obtain information on the process by which a program achieves its effects and whether the program is effective for subgroups of individuals. A general model that simultaneously estimates mediation and moderation effects is presented, and the utility of combining the effects into a single model is described. Possible effects of interest in the model are explained, as are statistical methods to assess these effects. The methods are further illustrated in a hypothetical prevention program example.

R2 effect-size measures for mediation analysis

R(2) effect-size measures are presented to assess variance accounted for in mediation models. The measures offer a means to evaluate both component paths and the overall mediated effect in mediation models. Statistical simulation results indicate acceptable bias across varying parameter and sample-size combinations. The measures are applied to a real-world example using data from a team-based health promotion program to improve the nutrition and exercise habits of firefighters. SAS and SPSS computer code are also provided for researchers to compute the measures in their own data.

Current Directions in Mediation Analysis

Mediating variables continue to play an important role in psychological theory and research. A mediating variable transmits the effect of an antecedent variable on to a dependent variable, thereby providing more detailed understanding of relations among variables. Methods to assess mediation have been an active area of research for the last two decades. This paper describes the current state of methods to investigate mediating variables.

Mediation analysis

Mediating variables are prominent in psychological theory and research. A mediating variable transmits the effect of an independent variable on a dependent variable. Differences between mediating variables and confounders, moderators, and covariates are outlined. Statistical methods to assess mediation and modern comprehensive approaches are described. Future directions for mediation analysis are discussed.

Mediation analysis

Mediating variables are prominent in psychological theory and research. A mediating variable transmits the effect of an independent variable on a dependent variable. Differences between mediating variables and confounders, moderators, and covariates are outlined. Statistical methods to assess mediation and modern comprehensive approaches are described. Future directions for mediation analysis are discussed.